Mammalian macrophage CSF (MCSF; CSF-1) is the primary regulator of the mononuclear phagocyte lineage. We, for the first time, report the complete sequencing of five MCSF cDNAs from three fish species, rainbow trout, zebrafish, and goldfish. Despite the difference in the lengths of the MCSF transcripts, all of the fish MCSF molecules encode a signal peptide, a CSF-1 domain, a transmembrane domain, and an intracellular region. Each fish MCSF gene has a unique exon/intron structure. The primordial MCSF gene may have had a nine exon/eight intron structure. In this model, insertion of an intron in exon 6 in primitive fish created the fish type I MCSF, while the loss of this exon or part of the original exon 6 created the fish type II MCSF. Investigation of alternative splicing variants in trout suggests that no mammalian equivalent splice variants exist. The two trout MCSF genes are differentially expressed in vivo and contributed differently to the high-level expression of MCSF in spleen and head kidney. In contrast to the up-regulation of MCSF by PMA in mammals, in trout MCSF1 expression is down-regulated by PMA treatment. As in mammals, recombinant trout MCSF1 can promote the growth of head kidney leukocytes, and it up-regulates the expression of CXCR3 in head kidney macrophages, with the latter suggesting a role of MCSF in the trafficking of macrophages to sites of inflammation or injury where the CXCR3 ligands are expressed. Thus MCSF has an important role in the immune system of fish as in mammals.

Macrophage CSF (MCSF),4 also known as CSF-1, is the primary regulator of the mononuclear phagocyte lineage, including monoblasts, promonocytes, monocytes, tissue macrophages, dendritic cells, microglia, and osteoclasts (1, 2, 3, 4). It has also been demonstrated to play important roles in bone metabolism, inflammation, pregnancy, and in preimplantation development of the female reproductive tract (5, 6). All effects of MCSF are mediated by a single high-affinity transmembrane receptor (MCSFR) encoded by the protooncogene c-fms (7).

Mammalian MCSF has three isoforms, a secreted glycoprotein, a secreted proteoglycan, and a biologically active, membrane-spanning, cell surface glycoprotein (8). Secreted and cell surface isoforms of CSF-1 can have different specific roles on target cells bearing MCSFR at their surface and differential effects in inflammation and immunity (4). All three isoforms are encoded by a single gene that has 10 exons and nine introns, spanning 21 kb (9). At least five mature human (and a similar number in mouse) MCSF mRNAs have been isolated that result from alternative splicing in exon 6 and the alternative usage of exon 9 and exon 10 in the 3′-untranslated region (UTR). Thus, three different classes of MCSF precursor proteins (256, 554, and 438 aa in length, respectively), designated MCSF-α, -β, and -γ, were predicted to be encoded by these mRNAs (10). The bioactive mammalian MCSF is a homodimer stabilized by intermolecular and intramolecular disulfide bonds (11, 12) that are crucial for attaining full biological activity.

Mammalian MCSF is produced in vitro by numerous cell lines, either constitutively (at low levels) or after induction. Proinflammatory stimuli, such as LPS, IFN-γ, and IL-1β induce MCSF production by monocytes, endothelial cells, fibroblasts, T cells, and polymorphonuclear leukocytes (13, 14, 15). PMA up-regulates mammalian MCSF expression in macrophages (16), B lymphocytes (17), T lymphocytes (18), and malignant cells (19). MCSF, by binding to its receptor, initiates receptor dimerization, autophosphorylation, and activation of its tyrosine kinase activity. Phosphorylated cytoplasmic secondary substrates induce a cascade of biochemical events leading to the following cellular responses: mitosis, secretion of cytokines, membrane ruffling, and autoregulation of transcription (i.e., MCSFR) (2, 20).

From an evolutionary standpoint, the mononuclear/phagocyte lineage, especially macrophages, and their function are well conserved. Macrophage-like cells can be found in almost all multicellular organisms, and fish macrophage development has been studied recently in zebrafish (21), goldfish (22), and medaka (23). Despite the recent identification of MCSFR in fish (24, 25, 26, 27), the growth factor of macrophages, MCSF, had not been reported in nonmammals until an incomplete MCSF molecule was described in goldfish (28). In this study, we describe the complete sequencing of four MCSF cDNA sequences from two fish species, rainbow trout (Oncorhynchus mykiss) and zebrafish (Danio rerio), as well as a complete goldfish MCSF cDNA. Two MCSF genes have been identified in fish that differ in gene organization and are differentially expressed. The recombinant trout MCSF1 protein is shown to promote the growth of head kidney leukocytes and modulate the expression of the chemokine receptor CXCR3 in macrophages, suggesting that it has an important role in the immune system of fish as it does in mammals.

Four rainbow trout cell lines were used: the mononuclear/macrophage cell line RTS-11 (29), the gonad fibroid cell line RTG-2 (30), the liver fibroid cell line RTL (31), and the liver epithelial cell line CL-6 (generous gift from Dr. A. Benmansour, Institut National de la Recherche Agronomique, Jouy-en-Josas, France). All trout cells were grown at 20°C in L-15 medium supplemented with 30% FCS for RTS-11 cells and 10% FCS for all other cell lines.

A spleen SSH library from Aeromonas salmonicida subsp. salmonicida, strain MT 423-infected trout was constructed and random sequence analysis of SSH clones were as described previously (32). One SSH clone was identified with homology to mammalian MCSF. 3′-RACE and 5′-RACE were conducted using SMART cDNA prepared from spleen as described previously (32). 5′-RACE using primers TF-R1 and TF-R2 (Table I) amplified a 0.3-kb band that was cloned and sequenced. 3′-RACE, using forward primers TF-F1 and TF-F2, resulted in a 2.8-kb major product that was cloned. Sequencing of this product revealed that it contained the complete coding region and putative 3′-UTR without a canonical poly(A) signal, suggesting that the 3′-end sequencing had not been completed. An additional 3′-RACE using primers TF-F3/TF-F4 at the C-terminal of the coding region amplified a 1.5-kb product that extended the cDNA sequence to 3.4 kb, but again without a poly(A) signal. This situation could result from a secondary structure in the 3′-UTR that impeded full-length cDNA synthesis initiated from the bone fide poly(A) tail of the mRNA at the low reaction temperature (42°C) used for SMART cDNA synthesis. Thus, the cDNA synthesis may have initiated at a poly(A)-rich sequence, resulting in a shortened 3′-RACE product. To resolve this problem, a 3′-RACE-specific cDNA was synthesized from spleen RNA using the same oligo(dT) adaptor primer as that for SMART cDNA synthesis (Clontech) and SuperScript III reverse transcriptase (Invitrogen) at 55°C. 3′-RACE using primers TF-F3/TF-F5 amplified a 4-kb product that extended the cDNA sequence to 7.2 kb and now contained a canonical poly(A) signal (AATAAA) present 16 bp upstream of the poly(A) tail, indicating the completion of the full-length cDNA sequencing.

Table I.

Primers used for gene cloning and expression

GenePrimerSequences (5′ to 3′)Application
Trout MCSF-1 TF-F1 CAA CAT CCG GAC GTT GAT CGT TC 3′-Racing and RT-PCR 
 TF-F2 CCT TTC CTT CTC TTG CTC TTC ATC 3′-Racing 
 TF-F3 CAG AAA GGT CCA GTG AAG GCT C 3′-Racing 
 TF-F4 AGC CCT TAT CCA CAG TAC CAG CA 3′-Racing 
 TF-F5 CTG TCG CAG TAA GGT TTG TGT TGT AG 3′-Racing 
 TF-F6 CAG CAG CAT TGA GGA CAT TGA GC RT-PCR 
 TF-R1 GAG CAG AGT TCC TGT CTT TGG TGA AGA TG 5′-Racing 
 TF-R2 GGC TGG TTT GTG TGT GTT CAT CTG GC 5′-Racing 
 TF-R3 CCG TTG TCC GAT TTC CTG AC RT-PCR 
 TF-R4 GAT GAC ACC TTG TAT TGC TTT GGC TC RT-PCR 
 TF-R5 CAC AAG GGT TAA GTG AAG AGG TGT ACT G RT-PCR 
 TF-F CTG AGC CAA ACC ATC CTA GGA C Real time PCR 
 TF-R GGC TTG GAG TCT CTT CTT CTC AC Real time PCR 
 TF-EF CTA CCA GGC AGT TGC TGA CAC RT-PCR 
 TF-ER CTG AGC CTT CAC TGG ACC TTT C RT-PCR 
    
Trout MCSF-2 TF2-F1 CAT AGG TGA GCC AGA TGA CC 3′-Racing 
 TF2-F2 GTG CCA ACC CAC ATT CAG TG 3′-Racing 
 TF2-F CCT CCC TAC AGC ACT CTC TCT GAC TAC Real time PCR 
 TF2-R GGT CAG TAC TGT AGG ACA TCT TGT GTG T Real time PCR 
    
Trout MCSFR TFR-F CAC ACA TAG ATC TGG AGC AGA TC RT-PCR 
 TFR-R CTT CCC CCT CGT TCA CGT C RT-PCR 
    
Trout β-actin RTBA-EF CGA CCT CAC AGA CTA CCT GAT RT-PCR 
 RTBA-ER TGG ATA CCG CAA GAC TCC ATA C RT-PCR 
    
Trout GAPDH GPDH-F ATG TCA GAC CTC TGT GTT GG RT-PCR 
 GPDH-R TCC TCG ATG CCG AAG TTG TCG RT-PCR 
    
Trout EF-1α EF1α-EF CAA GGA TAT CCG TCG TGG CA Real time PCR 
 EF1α-ER ACA GCG AAA CGA CCA AGA GG Real time PCR 
    
Trout CXCR3 receptor CXCR3A-F GTT GGT AAC AAT GGA TCA CGT CAA GGC Real time PCR 
 CXCR3A-R CAC ACA CAG CAC CAG GAT GTT ACC CAC Real time PCR 
    
Trout IL-11 IL-11-F CTG CTC TCG CTG CTA TTG G Real time PCR 
 IL-11-R TGG GTC TCA TCT CAA GGG AGT Real time PCR 
    
Zebrafish MCSF-1 ZF1-F1 CTG ACT TCA TTA GGA ACA AAC GCA AGA C 3′-Racing 
 ZF1-F2 GAG GAA GTG ATT CGC ATG TAT ATG AGT C 3′-Racing 
 ZF1-R1 TCT CTG TGA GGA TGT TGT AGA TTT CAG GT 5′-Racing 
 ZF1-R2 CAA ATG CAG GGG CAC TCT GAG CTA C 5′-Racing 
    
Zebrafish MCSF-2 ZF2-F1 CAA CTG AGT TAT GAG CTT TGC TGT CAT TC 3′-Racing 
 ZF2-F2 GGT GAG CCA GAT GAA CAA CCC TAC AC 3′-Racing 
 ZF2-R1 CTG TAC TCT TGC TCG CAG GTC C 5′-Racing 
 ZF2-R2 TCT CGA ATG CTA CAG GGT CCT C 5′-Racing 
    
Goldfish MCSF GF-F1 CAC TCT GTC ACC CAG GAC CAT C 3′-Racing 
 GF-F2 CCT ACA CCT TCA CTG AGC AAC AGA ACC 3′-Racing 
    
Adaptor primers for RACE SPS AAG CAG TGG TAT CAA CGC AGA GTA CGC GGG 5′-RACE cDNA synthesis 
 SPC-T30 AAG CAG TGG TAT CAA CGC AGA GTA CTT TTT TTT TTT TTT TTT TTT TTT TTT TTT T(A/G/C)N 3′-RACE cDNA synthesis 
 HP GTC AGC TAC GTC TCC TCA GTC AAG CAG TGG TAT CAA CGC AGA GT 3′ or 5′-Racing 
 HPS GTC AGC TAC GTC TCC TCA GTC 3′ or 5′-Racing 
 HPS1 CCT CAG TCA AGC AGT GGT ATC 3′ or 5′-Racing 
GenePrimerSequences (5′ to 3′)Application
Trout MCSF-1 TF-F1 CAA CAT CCG GAC GTT GAT CGT TC 3′-Racing and RT-PCR 
 TF-F2 CCT TTC CTT CTC TTG CTC TTC ATC 3′-Racing 
 TF-F3 CAG AAA GGT CCA GTG AAG GCT C 3′-Racing 
 TF-F4 AGC CCT TAT CCA CAG TAC CAG CA 3′-Racing 
 TF-F5 CTG TCG CAG TAA GGT TTG TGT TGT AG 3′-Racing 
 TF-F6 CAG CAG CAT TGA GGA CAT TGA GC RT-PCR 
 TF-R1 GAG CAG AGT TCC TGT CTT TGG TGA AGA TG 5′-Racing 
 TF-R2 GGC TGG TTT GTG TGT GTT CAT CTG GC 5′-Racing 
 TF-R3 CCG TTG TCC GAT TTC CTG AC RT-PCR 
 TF-R4 GAT GAC ACC TTG TAT TGC TTT GGC TC RT-PCR 
 TF-R5 CAC AAG GGT TAA GTG AAG AGG TGT ACT G RT-PCR 
 TF-F CTG AGC CAA ACC ATC CTA GGA C Real time PCR 
 TF-R GGC TTG GAG TCT CTT CTT CTC AC Real time PCR 
 TF-EF CTA CCA GGC AGT TGC TGA CAC RT-PCR 
 TF-ER CTG AGC CTT CAC TGG ACC TTT C RT-PCR 
    
Trout MCSF-2 TF2-F1 CAT AGG TGA GCC AGA TGA CC 3′-Racing 
 TF2-F2 GTG CCA ACC CAC ATT CAG TG 3′-Racing 
 TF2-F CCT CCC TAC AGC ACT CTC TCT GAC TAC Real time PCR 
 TF2-R GGT CAG TAC TGT AGG ACA TCT TGT GTG T Real time PCR 
    
Trout MCSFR TFR-F CAC ACA TAG ATC TGG AGC AGA TC RT-PCR 
 TFR-R CTT CCC CCT CGT TCA CGT C RT-PCR 
    
Trout β-actin RTBA-EF CGA CCT CAC AGA CTA CCT GAT RT-PCR 
 RTBA-ER TGG ATA CCG CAA GAC TCC ATA C RT-PCR 
    
Trout GAPDH GPDH-F ATG TCA GAC CTC TGT GTT GG RT-PCR 
 GPDH-R TCC TCG ATG CCG AAG TTG TCG RT-PCR 
    
Trout EF-1α EF1α-EF CAA GGA TAT CCG TCG TGG CA Real time PCR 
 EF1α-ER ACA GCG AAA CGA CCA AGA GG Real time PCR 
    
Trout CXCR3 receptor CXCR3A-F GTT GGT AAC AAT GGA TCA CGT CAA GGC Real time PCR 
 CXCR3A-R CAC ACA CAG CAC CAG GAT GTT ACC CAC Real time PCR 
    
Trout IL-11 IL-11-F CTG CTC TCG CTG CTA TTG G Real time PCR 
 IL-11-R TGG GTC TCA TCT CAA GGG AGT Real time PCR 
    
Zebrafish MCSF-1 ZF1-F1 CTG ACT TCA TTA GGA ACA AAC GCA AGA C 3′-Racing 
 ZF1-F2 GAG GAA GTG ATT CGC ATG TAT ATG AGT C 3′-Racing 
 ZF1-R1 TCT CTG TGA GGA TGT TGT AGA TTT CAG GT 5′-Racing 
 ZF1-R2 CAA ATG CAG GGG CAC TCT GAG CTA C 5′-Racing 
    
Zebrafish MCSF-2 ZF2-F1 CAA CTG AGT TAT GAG CTT TGC TGT CAT TC 3′-Racing 
 ZF2-F2 GGT GAG CCA GAT GAA CAA CCC TAC AC 3′-Racing 
 ZF2-R1 CTG TAC TCT TGC TCG CAG GTC C 5′-Racing 
 ZF2-R2 TCT CGA ATG CTA CAG GGT CCT C 5′-Racing 
    
Goldfish MCSF GF-F1 CAC TCT GTC ACC CAG GAC CAT C 3′-Racing 
 GF-F2 CCT ACA CCT TCA CTG AGC AAC AGA ACC 3′-Racing 
    
Adaptor primers for RACE SPS AAG CAG TGG TAT CAA CGC AGA GTA CGC GGG 5′-RACE cDNA synthesis 
 SPC-T30 AAG CAG TGG TAT CAA CGC AGA GTA CTT TTT TTT TTT TTT TTT TTT TTT TTT TTT T(A/G/C)N 3′-RACE cDNA synthesis 
 HP GTC AGC TAC GTC TCC TCA GTC AAG CAG TGG TAT CAA CGC AGA GT 3′ or 5′-Racing 
 HPS GTC AGC TAC GTC TCC TCA GTC 3′ or 5′-Racing 
 HPS1 CCT CAG TCA AGC AGT GGT ATC 3′ or 5′-Racing 

An expressed sequence tag (EST) database search (www.ncbi.nlm.nih.gov/blast/Blast.cgi) with the cloned trout MCSF (referred to as MCSF1 thereafter) revealed a trout EST (GenBank accession no. CX250296) that, when translated, had homology to MCSF1. 3′-RACE was conducted using primers TF2-F1 and TF2-F2 (Table I) from head kidney cDNA as described above to obtain cDNA that contained the complete coding region and 3′-UTR.

The trout MCSF1 cDNA encodes a comparable peptide to that of mammalian MCSF with a transmembrane domain. A search of National Center for Biotechnology Information (NCBI) databases (www.ncbi.nlm.nih.gov/blast/Blast.cgi) with the trout molecule was unable to identify any MCSF molecules in chicken and Xenopus but gave hits to known mammalian genes and a few in other fish. A predicted zebrafish MCSF-like molecule (GenBank accession no. XM_001343834) and a recently reported goldfish MCSF molecule (28) only encodes a short peptide equivalent to the N terminus of the trout and mammalian MCSF molecules without a transmembrane domain. A second zebrafish MCSF molecule (GenBank accession no. NM_001080076) encoding a short peptide was also found. To clarify these molecules, we recloned the zebrafish MCSF molecules by RACE using fin total RNA, as our initial analysis of MCSF expression in zebrafish tissues suggested that fins expressed a high level of MCSF. The 5′- and 3′-RACE cDNAs were synthesized using PowerScript (Clontech) and SuperScript III reverse transcriptase (Invitrogen), respectively, as described above. We also identified a stickleback genomic sequence (GenBank accession no. AANH01010606) that has a similar gene organization to trout and zebrafish MCSF2 genes.

It was possible that technical issues had affected the previous cloning of the 3′-end of goldfish MCSF cDNA (28), because both the zebrafish gene and the trout gene possessed a transmembrane domain. 3′-RACE was performed using primer GF-F1/GF-F2 (Table I) and goldfish spleen cDNA synthesized at an elevated temperature as described above. A 2.2-kb product was obtained and, when fully sequenced, was found to contain the rest of the coding region and the 3′-UTR.

The nucleotide sequences generated were assembled and analyzed with the AlignIR program (LI-COR). Homology searching was conducted using the basic local alignment search tool (BLAST) program (www.ncbi.nlm.nih. gov/BLAST) (33). Protein identification was conducted using the Expert Protein Analysis System (ExPASy; www.expasy.org/tools/) (34), the signal peptide was predicted using the SignalP 3.0 program (35) and the transmembrane domain was predicted using the TMpred program (36). Global sequence comparison was performed using the MatGAT program (37). Protein modular structure was predicted using the SMART program (38). Multiple sequence alignments were generated using ClustalW (version 1.82) (39) and shaded using BOXSHADE (version 3.21; www.ch. embnet.org/software/BOX_form.html). Synteny analysis of orthologs between human and zebrafish was conducted using Cinteny (40), with the databases Homo sapiens Build 36.2 and D. rerio Zv6. Phylogenetic trees were created by the neighbor-joining method using the MEGA program (version 4.1) (41) and were bootstrapped 10,000 times.

Analysis of five overlapping PCR products amplified from trout genomic Lambda DNA sublibraries (42) resulted in a contiguous 26.6-kb genomic sequence that contained the trout MCSF gene (25,881 bp) and 3′-flanking region (656 bp). The genomic sequences of zebrafish MCSF1 (GenBank accession no. NW_001510630) and MCSF2 (GenBank accession no. NW_001513985) were extracted from the database. The resulting genomic sequence was aligned with the obtained cDNA sequence using the SIM4 program (43) and the exon/intron boundaries identified. To examine whether any alternative splicing variants exist for the trout MCSF1 gene, different primer pairs across different exons were used for PCR to amplify cDNA from different tissue samples, and the resulting products were differentiated on agarose gels as described previously (42).

RNA preparation, cDNA synthesis, and real-time PCR analysis of gene expression were as described previously (44, 45). Six healthy rainbow trout (average weight, 120 g) were killed and eight tissues (gills, skin, muscle, liver, spleen, head kidney, intestine, and brain) were collected. To directly compare the expression level of the two trout MCSF genes, a reference was constructed using equal molar amounts of PCR product from each gene. The expression level of each gene in different tissues was expressed as arbitrary units that were normalized against the expression level of EF-1α. The contribution of each gene to MCSF expression (relative expression) in specific tissues was calculated as a percentage at the molar level.

A quantity of 2 × 106 cells (RTG-2, RTL or CL-5) or 3 × 106 cells (RTS-11) was seeded into 25-m (2) culture flasks in 5 ml of complete medium (L-15 plus 10% FCS) and cultured at 20°C overnight before any treatments or RNA preparation. Escherichia coli LPS (25 μg/ml; Sigma-Aldrich) and polyinosinic:polycytidylic acid (poly(I:C)) (50 μg/ml; Sigma-Aldrich) was initially added to all cell lines for 4 h to investigate their ability to modulate the expression of MCSF. Total RNA was prepared from cells and reverse transcribed into cDNA as described previously (45); the primers used in PCRs for the expression of β-actin (forward and reverse), GAPDH (forward and reverse), MCSF1 (EF2 and ER2), and its receptor MCSF-R (forward and reverse) are shown in Table I. PMA treatment experiments were conducted in a similar way using a dose range of 0.5–500 ng/ml PMA and timings up to 24 h. The expression of MCSF1 and MCSF2, as well that of as the control genes EF-1α and IL-11, was detected by real-time PCR using three individual flasks of cells for each treatment. IL-11 (46) was studied because it is known to be up-regulated by PMA in these cell lines and thus can confirm that any negative effects on MCSF expression are not due to toxic effects.

The full-length trout MCSF1 extracellular domain was used to produce a recombinant protein for bioactivity analysis. Briefly, the trout MCSF sequence encoding the extracellular domain (the first 524 aa) was amplified using primers (forward: 5′-CACCGCCATGAACACACACAAACCAGC-3′; reverse: 5′-CCCGTGTCGGAATTCCTTCAC-3′) and directly cloned to pcDNA 6.2/GW/D-TOPO (Invitrogen). To facilitate purification and detection of the produced recombinant protein, the V5 tag in the vector was replaced at the Csp45I site by a Lumio-His6 tag from the pET161 vector (Invitrogen). Thus, the rtMCSF construct encodes for the 29-aa signal peptide, the first 495-aa mature peptide, and a 44-aa C-terminal Lumio-His6 tag (GGRADPAFLYKVVDLEGPRFAKLEAGGCCPGCCGGGTGHH HHHH). A sequence-confirmed plasmid was linearized using ScaI and used to transform CHO cells by electroporation using a Multiporator (Eppendorf). The transformed cells were selected using 5 μg/ml blasticidin (Invitrogen). The cell culture supernatant and cell lysate were used for rMCSF purification using a His GraviTrap affinity column and a His buffer kit (GE Healthcare). The purification of the rtMCSF was analyzed using electrophoresis with 4–12% SDS-polyacrylamide gels and detected first with the Lumio reagent (Invitrogen) and subsequently by staining with Brilliant Blue R (Sigma-Aldrich).

To produce rMCSF in E. coli, the MCSF1 sequence without the signal peptide but including the Lumio-His6 tag from the construct above was transferred to the XhoI/PvuII sites of the pBAD/gIII A vector (Invitrogen) using forward (5′-GCCTCGAG (XhoI site underlined) GGCGTCCCTGGTCCATGT-3′) and reverse (5′-CTGTGTTAGCAGCCGATCAAACTC-3′) primers. The construct was used to transform E. coli Top-10 cells (Invitrogen), after which plasmid DNA was prepared and sequenced. A sequence-confirmed Top-10 clone was directly used for rMCSF production. Briefly, a single colony was inoculated into 100 ml of Luria-Bertani buffer containing 100 μg/ml carbenicillin (Sigma-Aldrich) at 220 rpm at 37°C overnight. After overnight culture, the bacteria were diluted 1/100 in fresh Luria-Bertani medium and incubated for a further 1.5 h before a 4-h induction of expression by 0.004% arabinose. The rMCSF produced in E. coli was purified as described above. Potential contamination of LPS was reduced by repeated passing of the purified rMCSF through a polymyxin B column (Sigma-Aldrich).

Trout head kidney leukocytes were isolated as described previously by Stafford et al. (47). The head kidney leukocytes were isolated by passing the kidney through a sterile nylon mesh screen and then centrifuging the cells at 200 × g for 10 min on a 51% Percoll gradient. Cells at the medium-Percoll interface were removed and washed twice in serum-free medium (48) before culture.

The WST-1 (tetrazolium salt 4-[3-(4(iodophenyl)-2-(4-nitrophenyl)-2H-5-tetrazole]-1,3-benzene disulfonate) cellular proliferation assay (Roche Diagnostics) was used to measure the effects of rtMCSF on trout macrophages. The assay was performed according to the manufacturer’s specifications and is based on cleavage of the tetrazolium salt into formazan by mitochondrial dehydrogenase activity. The isolated head kidney cells were counted using a hemocytometer and added to a 96-well plate at 104 or 105 cells/well in 50 μl. Head kidney leukocytes were obtained from individual fish for proliferation determination for a complete set of treatment conditions. Eight independent experiments were performed. The experimental groups included one of the following: rtMCSF1 (40 ng), trout cell-conditioned medium (47), the elution buffer used to elute rtMCSF1 from the nickel column, transfected CHO cell lysates containing rtMCSF1, untransfected CHO cell lysates, or cell culture medium only. The total volume in each well was 100 μl. The treatments were performed immediately after cell isolation from the head kidney and every 48 h thereafter until the end of the experiments. The proliferation assay was performed every 2 days until day 10 of cultivation. At each time point, 10 μl of WST-1 was added to each well and allowed to incubate at 20°C for 4 h before measuring the absorbance at 450 nm using a microplate reader (BioTek Instruments).

rtMCSF1 prepared from E. coli was added to freshly prepared or cultured macrophages. The heat-inactivated (98°C, 30 min) rtMCSF preparation was used as a control. The expression of CXCR3 and EF-1α was detected by real-time RT-PCR as above. The expression of CXCR3 and the macrophage marker gene MCSFR was also examined in in vitro cultured adherent (macrophage-enriched) and nonadherent head kidney leukocytes.

The cloning and major features of the five MCSF cDNAs from three fish species are detailed in Fig. 1 A. The contiguous cDNA sequences of trout MCSF1, MCSF2, zebrafish MCSF1, zebrafish MCSF2, and goldfish MCSF are 7,168, 1,733, 2,460, 1,472, and 2,484 bp, respectively, with a poly(A) signal (AATAAA or ATTAAA) followed by a poly(A) tail. Multiple upstream ATG codons in the 5′-UTRs were found in all the fish MCSF cDNAs (three in trout and zebrafish MCSF1s, four in zebrafish MCSF2, two in goldfish MCSF, and at least one in trout MCSF2). The trout MCSF1 cDNA has a large 3′-UTR of 5,090 bp that contains five poly(A) stretches (more than seven A nucleotides), and ten mRNA instability motifs (ATTTA). One poly(A) stretch is also found in zebrafish MCSF1 and goldfish MCSF. There are also two ATTTA motifs in the 3′-UTR of zebrafish MCSF1 and trout MCSF2 and one in goldfish MCSF, but none in zebrafish MCSF2. A stem-loop could also be formed between C3190-C3203 and G4517-G4504 that could impede the full-length cDNA synthesis from the authentic poly(A) tail of the trout MCSF1 mRNA. The trout MCSF1 and 2, zebrafish MCSF1 and 2, and goldfish MCSF have open reading frames encoding for 593, 276, 526, 284, and 544 aa, respectively.

FIGURE 1.

The cloning strategy and major features of fish MCSF cDNA molecules (A) and the modular structure of MCSF molecules (B). A, Schematic diagram of the cloning and major features of fish MCSF cDNAs. 3′-RACE and 5′-RACE were used to isolate fish MCSF cDNAs. Note that three 3′-RACEs were performed to obtain the full-length trout MCSF1 cDNA, where the third 3′-RACE was performed on cDNA synthesized at 55°C. The primer and RACE directions are indicated by arrows. The trout MCSF1 transcript contains a large 3′-UTR of 5 kb containing five poly(A) stretches (more than seven poly(A) codons (7As)), 10 ATTTA mRNA instability motifs, and two 14-bp reverse repeat sequences that could form a stable loop secondary structure. B, The modular structures of fish and human MCSF molecules were predicted using the SMART program. The lengths of proteins are on the right. Note tht all proteins have a signal peptide, a CSF-1 domain, a transmembrane domain separated by a linker region of different lengths that may harbor segments of low compositional complexity, and an intracellular region.

FIGURE 1.

The cloning strategy and major features of fish MCSF cDNA molecules (A) and the modular structure of MCSF molecules (B). A, Schematic diagram of the cloning and major features of fish MCSF cDNAs. 3′-RACE and 5′-RACE were used to isolate fish MCSF cDNAs. Note that three 3′-RACEs were performed to obtain the full-length trout MCSF1 cDNA, where the third 3′-RACE was performed on cDNA synthesized at 55°C. The primer and RACE directions are indicated by arrows. The trout MCSF1 transcript contains a large 3′-UTR of 5 kb containing five poly(A) stretches (more than seven poly(A) codons (7As)), 10 ATTTA mRNA instability motifs, and two 14-bp reverse repeat sequences that could form a stable loop secondary structure. B, The modular structures of fish and human MCSF molecules were predicted using the SMART program. The lengths of proteins are on the right. Note tht all proteins have a signal peptide, a CSF-1 domain, a transmembrane domain separated by a linker region of different lengths that may harbor segments of low compositional complexity, and an intracellular region.

Close modal

The full-length translations of trout MCSF1 share relatively higher identities (33–34%) to those of zebrafish MCSF1 and goldfish MCSF, but lower identities (19–22%) to trout and zebrafish MCSF2s and the mammalian MCSF molecules (Table II). The trout MCSF2 shares 44% identity with zebrafish MCSF2 but shares low homology with other fish and mammalian MCSF molecules (15–24% identities; Table II). However, the homology between the N termini of fish MCSF molecules encoded by the first five exons are high, showing 46–88% identities between fish molecules. However, the homologies of the N termini between fish and mammalian molecules are still low, with only 20–26% identities (Table II).

Table II.

Identity(%)/similarity(%) of the full-length (top right) and N termini encoded by exons 1–5 (bottom left) of MCSF molecules from fish and mammalsa

N Terminus Encoded by Exons 2–59
12345678
1. OmMCSF-1  21.6/29.8 33.9/49.1 18.7/28.0 33.4/48.6 22.3/38.1 21.2/37.9 21.3/37.3 20.8/36.6 
2. OmMCSF-2 49.5/65.5  24.4/33.1 44.4/64.1 23.0/30.9 15.6/25.8 15.4/24.2 15.1/27.9 14.8/24.2 
3. DrMCSF-1 67.0/78.0 49.7/64.4  21.8/32.1 67.8/78.5 21.5/37.4 21.2/39.7 19.3/38.8 19.4/39.0 
4. DrMCSF-2 44.6/66.0 54.4/75.8 48.4/69.6  19.8/30.3 15.1/23.3 13.9/22.9 14.7/24.3 13.2/22.3 
5. CaMCSF 63.4/74.3 48.7/64.4 88.4/92.6 46.4/68.6  21.1/40.8 23.4/42.6 20.3/43.1 20.3/41.0 
6. HsMCSF 23.6/43.5 20.0/44.3 23.3/44.7 23.7/46.6 26.2/45.8  75.3/81.4 70.6/80.0 62.7/76.0 
7. TmCSF 23.2/42.9 19.5/44.8 21.4/44.2 22.1/45.5 23.5/45.3 80.7/90.6  66.3/75.8 59.2/71.7 
8. MmMCSF 22.8/43.5 20.0/42.3 24.6/46.3 23.6/47.1 25.1/45.8 86.2/90.6 81.2/88.4  80.7/87.6 
9. RnMCSF 23.3/41.9 20.4/42.3 21.9/42.1 22.7/45.0 22.1/42.6 73.5/89.0 85.1/94.5 74.6/85.6  
N Terminus Encoded by Exons 2–59
12345678
1. OmMCSF-1  21.6/29.8 33.9/49.1 18.7/28.0 33.4/48.6 22.3/38.1 21.2/37.9 21.3/37.3 20.8/36.6 
2. OmMCSF-2 49.5/65.5  24.4/33.1 44.4/64.1 23.0/30.9 15.6/25.8 15.4/24.2 15.1/27.9 14.8/24.2 
3. DrMCSF-1 67.0/78.0 49.7/64.4  21.8/32.1 67.8/78.5 21.5/37.4 21.2/39.7 19.3/38.8 19.4/39.0 
4. DrMCSF-2 44.6/66.0 54.4/75.8 48.4/69.6  19.8/30.3 15.1/23.3 13.9/22.9 14.7/24.3 13.2/22.3 
5. CaMCSF 63.4/74.3 48.7/64.4 88.4/92.6 46.4/68.6  21.1/40.8 23.4/42.6 20.3/43.1 20.3/41.0 
6. HsMCSF 23.6/43.5 20.0/44.3 23.3/44.7 23.7/46.6 26.2/45.8  75.3/81.4 70.6/80.0 62.7/76.0 
7. TmCSF 23.2/42.9 19.5/44.8 21.4/44.2 22.1/45.5 23.5/45.3 80.7/90.6  66.3/75.8 59.2/71.7 
8. MmMCSF 22.8/43.5 20.0/42.3 24.6/46.3 23.6/47.1 25.1/45.8 86.2/90.6 81.2/88.4  80.7/87.6 
9. RnMCSF 23.3/41.9 20.4/42.3 21.9/42.1 22.7/45.0 22.1/42.6 73.5/89.0 85.1/94.5 74.6/85.6  
a

Note that the N terminus of fish MCSF molecules shares higher homology. Area in italics represent full-length protein, and areas in bold represent fish comparisons. Om, Oncorhynchus mykiss; Dr, Danio rerio; Ca, Carassius auratus; Hs, Homo sapiens; Mm, Mus musculus; Bt, Bos taurus; Rn, Rattus norvegicus.

Despite the differences in size of the translated proteins, where even lower homology between fish and mammalian molecules is seen, all of the MCSF molecules have a similar modular structure, including a signal peptide, a CSF-1 domain, a transmembrane domain, and an intracellular region (Fig. 1,B). The differences in protein size are caused by variations in the distances between the CSF-1 domain and transmembrane domain in different molecules where segments of low compositional complexity may exist. A multiple alignment (Fig. 2) shows the conservation of critical residues important for mammalian MCSF function (9, 11), including the cysteine residues C7, C48, C90, and C139 (numbered according to human MCSF) for intrachain disulfide bonds and the histidine residues H9 and H15 involved in interaction with the human MCSFR (10). The two cysteines (C157 and C159) forming interchain disulfide bridges are also preserved in trout and zebrafish MCSF1s but are missing in trout and zebrafish MCSF2s. The C31 forming an interchain disulfide bond in mammalian MCSF molecules is preserved in zebrafish MCSF1 and trout and zebrafish MCSF2s but is missing in trout MCSF1.

FIGURE 2.

Multiple alignment of the predicted fish MCSF translations with selected mammalian MCSF molecules. Identical (black background) and similar (gray background) residues identified by the ClustalW program are indicated. A boxed “X” representing the intron positions or a boxed “dash” (-) representing an intron missing is inserted before the amino acids encoded by the next exon (phase 0 intron) or before (phase 1 intron) or after (phase 2 intron) the amino acids that cross an intron. The signal peptide and transmembrane region are indicated above the alignment, as are the conserved cysteine and histidine residues (in boldface), and the cysteine residues with limited conservation (in boldface with shading). Note that the cysteine and histidine residues are numbered according to the human MCSF molecule. Om, Oncorhynchus mykiss; Dr, Danio rerio; Hs, Homo sapiens; Bt, Bos taurus; Mm, Mus musculus; Rn, Rattus norvegicus.

FIGURE 2.

Multiple alignment of the predicted fish MCSF translations with selected mammalian MCSF molecules. Identical (black background) and similar (gray background) residues identified by the ClustalW program are indicated. A boxed “X” representing the intron positions or a boxed “dash” (-) representing an intron missing is inserted before the amino acids encoded by the next exon (phase 0 intron) or before (phase 1 intron) or after (phase 2 intron) the amino acids that cross an intron. The signal peptide and transmembrane region are indicated above the alignment, as are the conserved cysteine and histidine residues (in boldface), and the cysteine residues with limited conservation (in boldface with shading). Note that the cysteine and histidine residues are numbered according to the human MCSF molecule. Om, Oncorhynchus mykiss; Dr, Danio rerio; Hs, Homo sapiens; Bt, Bos taurus; Mm, Mus musculus; Rn, Rattus norvegicus.

Close modal

The zebrafish MCSF1 gene is located at 20.6 megabases (Mb) on chromosome (Ch) 11, and the MCSF2 gene is located at 47.3 Mb on Ch 8 (D. rerio Zv6). A synteny analysis between zebrafish Ch 8 and 11 and human Ch 1, where human MCSF is located, is presented in Fig. 3. There are multiple marker genes, including MCSF, ATXN7L2, and TIMM17 on all the three chromosomes. SYT6, IL-10, and a block of three genes (FOXJ3, SORT1, and PSMA5) only exist on zebrafish Ch 11 and human Ch 1. GNAT2, MAGI3, and another block of three genes (TRIM33, BACS2, and DENND2C) are only present on zebrafish Ch 8 and human Ch 1. RBM39 is only present on zebrafish Ch 8 and 11 without an ortholog on human Ch 1. This suggests that the chromosome regions harboring the two zebrafish MCSF paralogs and the human ortholog have a common origin and that the two zebrafish paralogs have resulted from a chromosome/genome duplication event.

FIGURE 3.

Diagram to show gene synteny at the MCSF locus in zebrafish Ch 8 and 11 and human Ch 1. The conserved genes are represented by blocked arrows, with the arrow showing the transcriptional direction. The gene name is shown below the arrow and its location on the chromosome (in Mb) is shown above the arrow. A block of three or more contiguous genes is boxed.

FIGURE 3.

Diagram to show gene synteny at the MCSF locus in zebrafish Ch 8 and 11 and human Ch 1. The conserved genes are represented by blocked arrows, with the arrow showing the transcriptional direction. The gene name is shown below the arrow and its location on the chromosome (in Mb) is shown above the arrow. A block of three or more contiguous genes is boxed.

Close modal

Sequencing of genomic DNA revealed that the trout MCSF1 gene has a 10 exon/nine intron structure that spans 25.9 kb (Fig. 4; also see Fig. 6). The zebrafish MCSF1 gene has a seven exon/six intron structure, but the trout and zebrafish MCSF2 genes have a nine exon/eight intron structure. Comparison of the exon/intron structure among the known MCSF genes revealed that the first five exons encoding for the signal peptide and CSF-1 domain from all of the species are well conserved, with four phase 0 introns. The linker regions between the CSF-1 domain and transmembrane domain in trout and zebrafish MCSF1 genes are encoded by a single exon, whereas in trout and zebrafish MCSF2s and mammalian MCSF genes this region is encoded by an exon that also encodes for the transmembrane domain. The zebrafish MCSF1 gene only has a single exon encoding for the transmembrane domain and intracellular region without any noncoding exons in the 3′-UTR. There are two exons encoding the intracellular region and one noncoding exon in the 3′-UTR in the two trout MCSF genes and in mammalian MCSF genes, except for those in humans, where two noncoding exons can be alternatively used but only one is present in a specific transcript. The zebrafish MCSF2 gene has a very small exon 8 apparently caused by deletion of the 3′ end of the exon, resulting in the loss of the stop codon and extending the open reading frame into the final noncoding exon. Overall, this suggests that a primordial MCSF gene was probably similar to the mammalian MCSF gene with a nine exon/eight intron structure with a big exon 6 encoding the linker and transmembrane domains and one noncoding exon for the 3′-UTR. After the fish-wide genome duplication, one of the duplicated fish MCSF genes appear to have had an intron gain within exon 6, producing a trout MCSF1 like the 10 exon/nine intron structure, whereas the other MCSF gene appears to have lost part of exon 6, giving rise to a fish-specific short MCSF molecule.

FIGURE 4.

Exon/intron structure of MCSF genes from fish and mammals. Exons, which are numbered according to the trout MCSF1 gene (bold numbers at the top of each column), are represented by a box with the numbers inside the box representing the size (bp). Introns are represented by a solid bar with the size (bp) above (in small numbers) and the intron phase below (in large bold numbers). The gene size (in kb) is on the far right. Note that the first five exons that encode the signal peptide and CSF-1 domains are conserved in terms of size of exons and intron phase. The data are extracted from Ensembl with the gene identifiers ENSG00000184371 (human), ENSMUSG00000014599 (mouse), and ENSBTAG00000000283 (cow) and from GenBank with accession numbers NW_001510630 (zebrafish MCSF1) and NW_001513985 (zebrafish MCSF2) and sequence data from this report.

FIGURE 4.

Exon/intron structure of MCSF genes from fish and mammals. Exons, which are numbered according to the trout MCSF1 gene (bold numbers at the top of each column), are represented by a box with the numbers inside the box representing the size (bp). Introns are represented by a solid bar with the size (bp) above (in small numbers) and the intron phase below (in large bold numbers). The gene size (in kb) is on the far right. Note that the first five exons that encode the signal peptide and CSF-1 domains are conserved in terms of size of exons and intron phase. The data are extracted from Ensembl with the gene identifiers ENSG00000184371 (human), ENSMUSG00000014599 (mouse), and ENSBTAG00000000283 (cow) and from GenBank with accession numbers NW_001510630 (zebrafish MCSF1) and NW_001513985 (zebrafish MCSF2) and sequence data from this report.

Close modal
FIGURE 6.

Alternative splicing of MCSF genes. Schematic illustration of the MCSF genes showing the exon/intron organizations and alternative splicing variants in rainbow trout and humans (A) and PCR evidence of alternative splicing in trout (B, C, and D). The noncoding exons in the 3′-end are indicated by white boxes. By alternative splicing of exon 6 and alternative use of either one of the 3′-noncoding exons, the human MCSF gene produces multiple transcripts that encode for three isoforms of MCSF. The black arrows indicate the transcription from gene to mRNA. No human equivalent splicing variants were detectable in the trout MCSF1 gene (B and C), except a minor one that uses a 92-bp downstream donor site in intron 1, as detected by primer pair F1+R3 (D). Note that the trout MCSF1 splicing variant has a premature stop codon and thus will not be translated into a bioactive protein with a CSF-1 domain.

FIGURE 6.

Alternative splicing of MCSF genes. Schematic illustration of the MCSF genes showing the exon/intron organizations and alternative splicing variants in rainbow trout and humans (A) and PCR evidence of alternative splicing in trout (B, C, and D). The noncoding exons in the 3′-end are indicated by white boxes. By alternative splicing of exon 6 and alternative use of either one of the 3′-noncoding exons, the human MCSF gene produces multiple transcripts that encode for three isoforms of MCSF. The black arrows indicate the transcription from gene to mRNA. No human equivalent splicing variants were detectable in the trout MCSF1 gene (B and C), except a minor one that uses a 92-bp downstream donor site in intron 1, as detected by primer pair F1+R3 (D). Note that the trout MCSF1 splicing variant has a premature stop codon and thus will not be translated into a bioactive protein with a CSF-1 domain.

Close modal

Further analysis of the EST and genome databases identified MCSF molecules in Atlantic salmon, common carp, Atlantic halibut, killifish, and stickleback, but no evidence of MCSFs in Xenopus and chicken. An unrooted phylogenetic tree revealed that two clades were formed among the fish MCSF molecules and separate from the mammalian MCSF clade (Fig. 5). Four fish MCSF molecules formed the fish MCSF type I clade, including the trout and zebrafish MCSF1 and the carp MCSF gene known to encode for longer MCSF molecules. Seven fish MCSF molecules formed the fish MCSF type II clade, including the complete trout and zebrafish sequences that encode shorter MCSF molecules. Both type I and type II MCSF molecules have been identified in zebrafish, trout, and salmon, suggesting that two MCSF genes likely exist in most fish species.

FIGURE 5.

An unrooted phylogenetic tree of the fish and mammalian MCSF molecules constructed using the neighbor-joining method within the MEGA program (41 ). Node values represent percent bootstrap confidence derived from 10,000 replicates. The GenBank accession numbers used in this study are trout (Oncorhynchus mykiss) omMCSF1, AJ555867; omMCSF2, AM949839; salmon (Salmo salar) ssMCSF1, DY713192; ssMCSF2, DW582275; zebrafish (Danio rerio) drMCSF1, AM901598; drMCSF2, AM901599; goldfish (Carassius auratus) caMCSF, AM982798; carp (Cyprinus carpio) ccMCSF, EX881752; killifish (Fundulus heteroclitus) fhMCSF, DR441520; halibut (Hippoglossus hippoglossus) hhMCSF, EB035272; stickleback (Gasterosteus aculeatus) gaMCSF, DT998450; human (Homo sapiens) hsMCSF, P09603; mouse (Mus musculus) mmMCSF, P07141; cow (Bos taurus) btMCSF, O77709; and rat (Rattus norvegicus) rnMCSF, Q8JZQ0.

FIGURE 5.

An unrooted phylogenetic tree of the fish and mammalian MCSF molecules constructed using the neighbor-joining method within the MEGA program (41 ). Node values represent percent bootstrap confidence derived from 10,000 replicates. The GenBank accession numbers used in this study are trout (Oncorhynchus mykiss) omMCSF1, AJ555867; omMCSF2, AM949839; salmon (Salmo salar) ssMCSF1, DY713192; ssMCSF2, DW582275; zebrafish (Danio rerio) drMCSF1, AM901598; drMCSF2, AM901599; goldfish (Carassius auratus) caMCSF, AM982798; carp (Cyprinus carpio) ccMCSF, EX881752; killifish (Fundulus heteroclitus) fhMCSF, DR441520; halibut (Hippoglossus hippoglossus) hhMCSF, EB035272; stickleback (Gasterosteus aculeatus) gaMCSF, DT998450; human (Homo sapiens) hsMCSF, P09603; mouse (Mus musculus) mmMCSF, P07141; cow (Bos taurus) btMCSF, O77709; and rat (Rattus norvegicus) rnMCSF, Q8JZQ0.

Close modal

By alternative splicing of exon 6 and alternative use of one of the two exons for the 3′-UTR, the human MCSF gene produces multiple transcripts with three different translations (Fig. 6). To examine whether any alternative splicing variants exist in the trout MCSF1 gene, which encodes a mature MCSF molecule with comparable length to that of mammalian MCSF molecules, different primer combinations across different exons were used for PCR with cDNAs from tissues and cell lines. There were no significant splicing variants detectable (Fig. 6, B and C) as primer pairs located in exon 1 and the beginning of exon 10, and exon 9 and the end of exon 10 could only amplify a major band of the expected size from spleen cDNA. However, one splicing variant was detected using a primer pair located in exon 1 and 4 (Fig. 6 D). Sequence analysis revealed that this splice variant used a cryptic donor site 92 bp downstream of the authentic donor site, resulting in a transcript with a premature stop of translation. There were no apparent splice variants found for trout MCSF2 (data not shown). Thus, the mammalian equivalent alternative splicing variants are not present in trout.

The transcript expression of the two trout MCSF genes in vivo was examined in eight tissues including gills, skin, muscle, liver, spleen, head kidney, intestine, and brain. The highest level of MCSF1 and MCSF2 expression was detected in spleen, and head kidney, respectively. Although brain, skin, and intestine expressed MCSF1 at a higher level, the gills and head kidney only expressed MCSF1 at a very low level (Fig. 7). MCSF2 is also highly expressed in skin, muscle, gills, spleen, and brain. Although spleen expressed the highest level of MCSF1 among all the tissues examined, the head kidney expressed the highest level of MCSF2 but the lowest level of MCSF1. Thus, while MCSF1 contributes the majority of MCSF expression in spleen, brain, and intestine, MCSF2 contributes the majority of MCSF expression in head kidney, gills, and muscle (Fig. 7). The skin and liver express both MCSF1 and MCSF2 at a more comparable level.

FIGURE 7.

In vivo expression of trout MCSF transcript. The expression of trout MCSF1 and MCSF2 transcripts in tissues of six fish was determined. The arbitrary units of each MCSF gene expression (top panel) in different tissues was normalized against the expression level of EF-1α. The relative expression of each gene in a specific tissue depicts its contribution to the percentage of the total MCSF expression. The results represent the average ± SEM of six fish.

FIGURE 7.

In vivo expression of trout MCSF transcript. The expression of trout MCSF1 and MCSF2 transcripts in tissues of six fish was determined. The arbitrary units of each MCSF gene expression (top panel) in different tissues was normalized against the expression level of EF-1α. The relative expression of each gene in a specific tissue depicts its contribution to the percentage of the total MCSF expression. The results represent the average ± SEM of six fish.

Close modal

The expression of MCSF1 was detectable in all of the cell lines examined, including the mononuclear cell line RTS-11 (29), the two fibroid cell lines RTG-2 (30) and RTL (31), and the epithelial cell line CL-6 (Fig. 8,A). MCSF1 expression in cell lines was not modulated by LPS and poly(I:C), two commonly used immune stimulants mimicking the infection of bacteria and viruses, respectively. The two fibroid cell lines RTG-2 and RTL express higher levels of MCSF1 relative to the other two cell lines. Similar expression results for MCSF2 have also been obtained in cell lines, although the expression of MCSF2 is significantly lower compared with that of MCSF1. Real-time PCR indicated that the expression of MCSF1 is 30- to 40-fold higher than that of MCSF2 in RTG-2 and RTS-11 cells (Fig. 8,D). The expression of the MCSF receptor, in contrast, is only detectable in the RTS-11 cells (Fig. 8,B). The expression of the MCSF1 and MCSF2 transcripts in trout cell lines can be up-regulated by IL-1β and IFN-γ, but only a maximum of a 2-fold change has been detected by real-time PCR using EF-1α as a reference (data not shown). In contrast to the PMA up-regulation of MCSF expression in mammals, PMA was found to down-regulate the expression of trout MCSF1. A detailed time course study in RTG-2 cells using 50 ng/ml PMA showed that MCSF1 expression was significantly decreased after 2 h of treatment with PMA and continued to decrease to 24 h, when the expression of MCSF1 was >10% of the control level (Fig. 8,C). The PMA inhibition of MCSF1 expression was seen with all of the doses tested (0.5 to 500 ng/ml), and as little as 0.5 ng/ml PMA could inhibit MCSF1 expression by >60% after 8 h treatment in both the RTG-2 and RTS-11 cell lines (Fig. 9,A). MCSF2 expression was also inhibited by PMA but to a lesser extent (data not shown). The inhibition of gene expression was not a common feature of PMA-induced effects in trout cell lines and, for example, PMA increased the expression of IL-11 in the same cell lines under the same conditions (Fig. 9 B).

FIGURE 8.

Expression and modulation of MCSF in cells lines. A, MCSF1 expression in four cell lines. Trout cell lines were untreated (lanes 1), stimulated with LPS (lanes 2) or poly(I:C) (lanes 3) for 4 h and total RNA was prepared. The expression of MCSF1 (EF+ER) as well as that of two house keeping genes (GAPDH and β-actin), was detected by RT-PCR. “Neg. Control” represents negative control. B, The expression of the receptor of MCSF, MCSFR, in different cell lines. Note that the expression of MCSFR is only detectable in the monocyte/macrophage cell line RTS-11. C, PMA down-regulation of the expression of MCSF1. RTG-2 cells were seeded at 2 × 106 cells/flask in 5 ml of complete medium, and at different times PMA was added to the supernatant at 50 ng/ml. All cultures were continued to 36 h when total RNA was prepared. The expression of MCSF1 was normalized to the expression level of EF-1α and compared with the expression level in the control cells where the expression of MCSF1 is defined as 100. The results represent the average ± SEM of three replicates. The experiments were repeated three times. D, Quantitative comparison of the expression of MCSF genes in RTS-11 and RTG-2 cell lines. Total RNA was prepared from 2-day old cells after passaging under normal cell culture conditions, and the expressions of MCSF1 and MCSF2 were determined. The values were normalized to the expression level of EF-1α and then divided by the average value of MCSF2 expression in RTS-11. The results represent the mean ± SEM of six replicates.

FIGURE 8.

Expression and modulation of MCSF in cells lines. A, MCSF1 expression in four cell lines. Trout cell lines were untreated (lanes 1), stimulated with LPS (lanes 2) or poly(I:C) (lanes 3) for 4 h and total RNA was prepared. The expression of MCSF1 (EF+ER) as well as that of two house keeping genes (GAPDH and β-actin), was detected by RT-PCR. “Neg. Control” represents negative control. B, The expression of the receptor of MCSF, MCSFR, in different cell lines. Note that the expression of MCSFR is only detectable in the monocyte/macrophage cell line RTS-11. C, PMA down-regulation of the expression of MCSF1. RTG-2 cells were seeded at 2 × 106 cells/flask in 5 ml of complete medium, and at different times PMA was added to the supernatant at 50 ng/ml. All cultures were continued to 36 h when total RNA was prepared. The expression of MCSF1 was normalized to the expression level of EF-1α and compared with the expression level in the control cells where the expression of MCSF1 is defined as 100. The results represent the average ± SEM of three replicates. The experiments were repeated three times. D, Quantitative comparison of the expression of MCSF genes in RTS-11 and RTG-2 cell lines. Total RNA was prepared from 2-day old cells after passaging under normal cell culture conditions, and the expressions of MCSF1 and MCSF2 were determined. The values were normalized to the expression level of EF-1α and then divided by the average value of MCSF2 expression in RTS-11. The results represent the mean ± SEM of six replicates.

Close modal
FIGURE 9.

Dose-response of the expression of MCSF1 (A) and IL-11 (B) to PMA treatment in RTG-2 and RTS-11 cells. RTG-2 cells (2 × 106) or RTS-11 cells (3 × 106) were seeded to 25-cm2 cell culture flasks in 5 ml of complete medium (L-15 with 10% FCS) and cultured overnight. PMA was added to the cells for 8 h at 20°C and total RNA was then prepared. The expression of MCSF1 was normalized to the expression level of EF-1α and compared with the expression level in the control cells where the expression of MCSF was defined as 100. The expression of IL-11 was also normalized to the expression level of EF-1α and the fold increase is given for each treatment compared with the expression level in the control cells where the expression of IL-11 was defined as 1. The results represent the mean ± SEM of three replicates. Experiments were repeated two times.

FIGURE 9.

Dose-response of the expression of MCSF1 (A) and IL-11 (B) to PMA treatment in RTG-2 and RTS-11 cells. RTG-2 cells (2 × 106) or RTS-11 cells (3 × 106) were seeded to 25-cm2 cell culture flasks in 5 ml of complete medium (L-15 with 10% FCS) and cultured overnight. PMA was added to the cells for 8 h at 20°C and total RNA was then prepared. The expression of MCSF1 was normalized to the expression level of EF-1α and compared with the expression level in the control cells where the expression of MCSF was defined as 100. The expression of IL-11 was also normalized to the expression level of EF-1α and the fold increase is given for each treatment compared with the expression level in the control cells where the expression of IL-11 was defined as 1. The results represent the mean ± SEM of three replicates. Experiments were repeated two times.

Close modal

The human rMCSF, containing the first 149 aa of the N-terminal region with a critical cysteine residue (C31) for interchain disulfide bond and dimer formation, is bioactive (11, 12). The trout MCSF1 has this cysteine residue missing, suggesting that the trout bioactive MCSF1 may need to be a longer sequence. Thus, the full-length trout MCSF1 extra-cellular domain was used to produce a recombinant protein for bioactivity analysis. The recombinant trout MCSF1 was initially produced and partially purified using a eukaryotic expression vector and transfection of CHO cells (Fig. 10,A). When the pCL/MCSF-transfected CHO lysate or partially purified rtMCSF1 was added to head kidney leukocytes, the leukocyte growth was enhanced significantly as measured using the WST-1 assay. Control cells treated with CHO cell lysate or elution buffer-treated (used for elution of rtMCSF1 from affinity column) cells had no significant difference in proliferation relative to the untreated controls (Fig. 10 B).

FIGURE 10.

Purification of rtMCSF1 from pCL/MCSF1-transfected CHO cell lysates (A) and growth promotion activities of rtMCSF1 on trout macrophages as determined by using the WST-1 assay (B). A, rtMCSF1 was purified from cell lysate using a His GraviTrap affinity column analyzed by electrophoresis using a 4–12% SDS-polyacrylamide gel and detected with a Lumio reagent. Lane 1, Cell lysate; lane 2, flow through; lanes 3 and 4, wash through; lanes 5–7, elutions; lane 8, protein marker. B, Proliferative response of trout macrophages treated with 50 μl of purified rtMCSF1 (MCSF1/purified), elution buffer, cell conditioned medium containing growth-inducing ability (CCM), pCL/MCSF1 transfected CHO cell lysates (CHO/MCSF1), control CHO cell lysates (CHO/Control), or trout macrophages only (Control). Each point on the graph is the mean ± SEM of n = 8 treatments except for the elution buffer group, where n = 4.

FIGURE 10.

Purification of rtMCSF1 from pCL/MCSF1-transfected CHO cell lysates (A) and growth promotion activities of rtMCSF1 on trout macrophages as determined by using the WST-1 assay (B). A, rtMCSF1 was purified from cell lysate using a His GraviTrap affinity column analyzed by electrophoresis using a 4–12% SDS-polyacrylamide gel and detected with a Lumio reagent. Lane 1, Cell lysate; lane 2, flow through; lanes 3 and 4, wash through; lanes 5–7, elutions; lane 8, protein marker. B, Proliferative response of trout macrophages treated with 50 μl of purified rtMCSF1 (MCSF1/purified), elution buffer, cell conditioned medium containing growth-inducing ability (CCM), pCL/MCSF1 transfected CHO cell lysates (CHO/MCSF1), control CHO cell lysates (CHO/Control), or trout macrophages only (Control). Each point on the graph is the mean ± SEM of n = 8 treatments except for the elution buffer group, where n = 4.

Close modal

To further investigate MCSF1 activity, the trout MCSF1 was expressed in E. coli and purified (data not shown). The purified protein was detoxified by repeat passing through a polymyxin column to remove any contaminating LPS. When the purified and detoxified rtMCSF1 was added to freshly prepared head kidney macrophage cell cultures and incubated overnight (20 h), it increased significantly the expression of the chemokine receptor CXCR3, whereas the heat-inactivated (98°C, 30 min) recombinant protein had no effect (Fig. 11,A). The expression of other genes, including IL-1β, IL-11, IL-15, and IFN-γ, was not modulated in head kidney leukocytes by rtMCSF1 treatment. rtMCSF1 also increased the expression of CXCR3 in in vitro cultured macrophages. There was a 5-fold increase of CXCR3 expression after a 4-h stimulation of overnight cultured head kidney leukocytes with rtMCSF1 (Fig. 11,B). These experiments were repeated at least three times. There were no effects detectable on the expression of IL-1β, TNF-α, IL-11, and IL-15 in rtMCSF1-stimulated head kidney leukocytes (data not shown). Mammalian CXCR3 is predominantly expressed in memory/activated T lymphocytes. Fish head kidney leukocytes contain a mixture of cell types, with T lymphocytes expected to reside in the nonadherent population. To verify whether the CXCR3 expression was within the adherent macrophages and/or the nonadherent head kidney leukocytes, cells were individually prepared from four healthy fish and the expression of CXCR3 and the macrophage marker gene MCSFR was examined after 5 and 44 h of in vitro culture. The 5-h time point was chosen because of the time needed for preparation of the adherent and nonadherent cells. The expression of MCSFR in adherent cells was >10-fold higher than that in the nonadherent cells after 5 h in vitro culture, while the expression of CXCR3 in adherent cells was 100-fold higher than in the nonadherent cells. After 44 h of in vitro culture, MCSFR expression was increased significantly (>50-fold) in the adherent cells, while in contrast a <2-fold increase was observed in the nonadherent cells. A 6-fold increase of the expression of CXCR3 was also observed in adherent cells after 44 h in vitro culture, but no apparent change of CXCR3 expression was seen in the nonadherent cells (Fig. 11 C). This suggests that the adherent macrophage is the major source of CXCR3 expression in head kidney cells. The expression of CXCR3 was also detectable in the RTS-11 monocyte/macrophage cell line (data not shown).

FIGURE 11.

RtMCSF1 modulation of CXCR3 expression in head kidney macrophages. A, Purified rtMCSF1 and heat-inactivated (98°C for 30 min) rtMCSF1 (rtMCSF1 heated) was added to freshly prepared head kidney macrophages at 200 ng/ml and incubated for 20 h. The expression of CXCR3 was detected by real-time RT-PCR and normalized to the expression level of EF-1α. The expression of CXCR3 is expressed as fold change compared with the untreated control (bar labeled “C”). The results represent the mean ± SEM of three fish. The experiments were repeated three times. B, Purified rtMCSF1 was added to overnight cultured macrophages at 200 ng/ml for 4 h and expression of CXCR3 was detected as described above. The results represent the average ± SEM of three fish. C, Expression of CXCR3 and MCSFR in adherent (A.) and nonadherent (N.A.) head kidney leukocytes. Head kidney leukocytes were resuspended in L-15 containing 0.1% FCS at 1 × 106 cells/ml, added to a cell culture flask, and incubated for 2 h at 20°C. The adherent cells were separated from nonadherent cells by repeated washing with L-15 with 0.1% FCS. Both cell types were cultured in L-15 with 10% FCS for 5 or 44 h before RNA preparation. The expression of CXCR3 and MCSFR was normalized to the expression level of EF-1α and compared with the expression level in the 5-h cultured adherent cells (macrophages) in which the expression level is defined as 100. The results represent the mean ± SEM of four fish.

FIGURE 11.

RtMCSF1 modulation of CXCR3 expression in head kidney macrophages. A, Purified rtMCSF1 and heat-inactivated (98°C for 30 min) rtMCSF1 (rtMCSF1 heated) was added to freshly prepared head kidney macrophages at 200 ng/ml and incubated for 20 h. The expression of CXCR3 was detected by real-time RT-PCR and normalized to the expression level of EF-1α. The expression of CXCR3 is expressed as fold change compared with the untreated control (bar labeled “C”). The results represent the mean ± SEM of three fish. The experiments were repeated three times. B, Purified rtMCSF1 was added to overnight cultured macrophages at 200 ng/ml for 4 h and expression of CXCR3 was detected as described above. The results represent the average ± SEM of three fish. C, Expression of CXCR3 and MCSFR in adherent (A.) and nonadherent (N.A.) head kidney leukocytes. Head kidney leukocytes were resuspended in L-15 containing 0.1% FCS at 1 × 106 cells/ml, added to a cell culture flask, and incubated for 2 h at 20°C. The adherent cells were separated from nonadherent cells by repeated washing with L-15 with 0.1% FCS. Both cell types were cultured in L-15 with 10% FCS for 5 or 44 h before RNA preparation. The expression of CXCR3 and MCSFR was normalized to the expression level of EF-1α and compared with the expression level in the 5-h cultured adherent cells (macrophages) in which the expression level is defined as 100. The results represent the mean ± SEM of four fish.

Close modal

In this study, we have first sequenced five full-length MCSF cDNA sequences in three fish species, for the first time in nonmammals. We have also analyzed the syntenic relationship and gene structure of MCSF genes and found that the two zebrafish MCSF paralogs appear to have arisen from a chromosome/genome duplication. We also investigated the expression and modulation of trout MCSF genes. Finally, we produced the trout recombinant MCSF1 and showed that it can promote the growth of head kidney leukocytes and up-regulate the expression of the chemokine receptor CXCR3.

Interestingly, in terms of their transcript and translation length, exon/intron structure, and gene size, each of the fish MCSF genes is unique. The trout MCSF1 gene has 10 exons and is transcribed into 7.2-kb mRNA encoding for 593 aa. The zebrafish MCSF1 gene only has seven exons and is transcribed into a 2.5-kb mRNA encoding for 526 aa. The trout and zebrafish MCSF2 genes have nine exons but are transcribed into short mRNAs (1.5–1.8 kb) encoding for short proteins (276 aa for trout MCSF2 and 284 aa for zebrafish MCSF2). Whereas the trout MCSF2 protein is encoded by the first eight exons, the zebrafish MCSF2 is encoded by all nine exons. Although the fish MCSF molecules share only limited homology with mammalian MCSF molecules (Table II), five lines of evidence indicate that they are indeed the orthologs of mammalian MCSF. First, both fish and mammalian MCSF molecules all share a similar modular structure of a signal peptide, a four-helical cytokine CSF-1 domain, a transmembrane domain separated by an unstructured linker, and an intracellular region. Second, the fish MCSF molecules share high homology between each other at the CSF-1 domain, which is important for MCSF function in mammals. They also share a relatively higher homology with mammalian counterparts at the CSF-1 domain (Table II). Third, critical cysteine residues responsible for disulfide bonds and histidine residues involved in receptor binding are conserved (Fig. 2). Fourth, the regions of the two chromosomes harboring the two zebrafish MCSF genes are syntenically related to the region harboring human MCSF on human chromosome 1 (Fig. 3). The exon/intron organization also shares some conservation between the fish and mammalian MCSF genes (Fig. 4). Lastly, trout rMCSF is a growth factor for head kidney macrophages (Fig. 10).

Two paralogous MCSF genes have been identified that reside on zebrafish CH11 and CH9, respectively. Syntenic analysis clearly showed the relationship of these two paralogous chromosomal regions with a region on human CH1 harboring the human MCSF gene (Fig. 4). This suggests that the two zebrafish MCSF paralogs have arisen from a chromosome or whole genome duplication event, coinciding with the whole genome duplication known to have occurred in the teleost fish lineage (49). It is expected that after a whole genome duplication the resulting polyploid genome gradually returns to a diploid state through extensive gene deletion. Paralogous chromosomes will thus each retain only a small subset of their initial common gene complement, and these will be broken into smaller segments by interchromosomal and intrachromosomal rearrangements. This is exemplified in the current study by the fact that some of conserved syntenic genes on human CH1 are preserved only on zebrafish CH8 or CH11. Two MCSF paralogs have also been identified in trout and salmon and may also exist in other teleosts as the result of the whole genome duplication event.

Two MCSF genes have been isolated from both zebrafish and trout in this study. Both trout and zebrafish MCSF2 genes have a nine exon/eight intron structure as is found in mammals except for primates, who have 10 exons with the last two being noncoding exons containing the 3′-UTR that are alternatively used (Figs. 4, 6). The alternative usage of two 3′-noncoding exons might be unique and only present in primates, as there is no evidence of this happening in other mammals including mouse, rat, and cow (data not shown). The trout MCSF1 gene has a similar gene organization as that in mammals, with complete conservation of the CSF-1 domain exon sizes and intron phase. However, the exception is the presence of an extra phase I intron 6 that gives a 10 exon/nine intron structure. The zebrafish MCSF1 gene also has an extra phase I intron 6, but it has lost all the three introns at the 3′-end. Although the type II fish MCSF gene has a conserved nine exon/eight intron structure, it is clear that exon 6 is dramatically reduced relative to that in mammals. There are two possibilities that could account for this if the primordial MCSF gene had a nine exon/eight intron structure with an big exon 6 encoding for the linker and transmembrane domains. In primitive fish after the fish-wide genome duplication event, it is possible that one of the MCSF genes gained a phase I intron (Fig. 4) in exon 6 and created a fish type I MCSF gene. The other duplicated MCSF gene appears to have either lost this new exon or lost part of the original exon 6 generating a fish type II gene.

The two trout MCSF genes were differentially expressed in vivo. Generally, MCSF gene expression is highly detectable in spleen, head kidney, skin, intestine, and gills, the major immune response-related tissues. It was also detectable in brain and muscle, suggesting a role outside the immune system. The kidney appears to be the primary location of hematopoiesis and monocyte/macrophage development in bony fishes. The spleen is another tissue rich in macrophages. Interestingly, the high level of MCSF expression in spleen is mainly contributed by the MCSF1 gene, whereas in head kidney the MCSF2 gene dominates (Fig. 7). These differential expression patterns of the two MCSF genes in trout may reflect in some way control mechanisms for macrophage development and activation in fish. We are producing the recombinant protein for the recently isolated trout MCSF2 to allow direct comparison of the function of the two fish MCSFs.

Mammalian MCSF expression is up-regulated by PMA in multiple systems (16, 17, 18, 19). In contrast, the expression of trout MCSF1 is down-regulated by PMA in both fibroid RTG-2 cells and monocyte/macrophage RTS-11 cells. This suggests that different mechanisms exist regarding the control of MCSF1 expression in trout and mammals. The fish MCSF genes all have multiple upstream ATG codons before the open reading frames (28) that could have a role in the control of protein translation (44). The trout MCSF1 gene also has a large 3′-UTR with multiple mRNA instability motifs and secondary structure (Fig. 1) that may impact on mRNA stability (50).

Unlike other cytokines, e.g., IL-1β (51) and IL-11 (46) where the homologies between orthologs from within fish groups and from fish to mammals are comparable, the homologies between the fish MCSF genes at the N-termini are considerably higher than between fish and mammals (46–84 vs 20–26% identities). This implies that the fish MCSF molecules may have novel functions compared with mammalian molecules. The observation that rtMCSF1 can promote the growth of head kidney macrophages (Fig. 10) was not surprising. However, we also found that rtMCSF1 up-regulates the expression of the chemokine receptor CXCR3 (Fig. 11), which is the receptor for CXCL9/MIG, CXCL10/IP-10, and CXCL11/I-TAC. It is predominantly expressed on memory/activated T lymphocytes in mammals and plays an important role in the recruitment of activated T cells to inflammatory sites (52). Interestingly, there is a single report that MCSF can up-regulate the expression of CXCR3 in mammalian osteoclasts where it is thought this helps recruit osteoclasts and their precursors to sites of bone remodeling (53). In trout, we show that CXCR3 appears to be preferentially expressed by adherent head kidney-derived macrophages as well as the fish monocyte/macrophage cell line RTS-11. The expression of CXCR3 on monocytes has also been described in grass carp recently (54). A CXCL9/CXCL10/CXCL11-like sequence has been identified in rainbow trout (55) that is up-regulated by poly(I:C) and trout IFN-γ (56). This suggests that fish MCSF may play a role in the trafficking and recruitment of macrophages to sites of infection by the up-regulation of CXCR3 expression.

We have sequenced five complete MCSF cDNAs from fish, the first for nonmammals. The two zebrafish MCSF paralogs appear to result from chromosome/whole genome duplication. The primordial MCSF may have had a nine exon/eight intron structure that was altered during evolution to give each fish MCSF gene a unique exon/intron structure. Insertion of an intron in exon 6 in primitive fish would have created the fish type I MCSF, whereas loss of this exon or part of the original exon 6 would give rise to the fish type II MCSF. The two trout MCSF genes are differentially expressed in vivo and contribute differently to the high level expression of MCSF in spleen and head kidney. The expression of trout MCSF1 is down-regulated by PMA. In addition to macrophage growth promotion activity, the trout MCSF1 up-regulates the expression of the chemokine receptor CXCR3, suggesting a novel role in macrophage trafficking.

Many thanks to Dr. Jun Zou (Scottish Fish Immunology Research Center, University of Aberdeen, Aberdeen, U.K.) for supplying the recombinant trout IL-1β and IFN-γ.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported financially by European Commission Contracts 513692 (Aquafirst), Q5RS-2001-002211 (Stressgenes), and 007103 (IMAQUANIM—Improved immunity of aquacultured animals).

2

The nucleotide sequences presented in this article have been submitted to the European Molecular Biology Laboratory/DNA Data Bank of Japan/GenBank nucleotide sequence database under the accession numbers AJ555867 (trout MCSF1 cDNA), AM901600 (trout MCSF1 DNA), AM949839 (trout MCSF2 cDNA), AM949840 (trout MCSF2 DNA), AM901598 (zebrafish MCSF1 cDNA), AM901599 (zebrafish MCSF2 cDNA), and AM982798 (goldfish MCSF cDNA).

4

Abbreviations used in this paper: MCSF, macrophage CSF; Ch, chromosome; CHO, Chinese hamster ovary; EST, expressed sequence tag; Mb, megabase; MCSFR, MCSF receptor; poly(I:C), polyinosinic:polycytidylic acid; rtMCSF, recombinant trout MCSF; UTR, untranslated region.

1
Hume, D. A., I. L. Ross, S. R. Himes, R. T. Sasmono, C. A. Wells, T. Ravasi.
2002
. The mononuclear phagocyte system revisited.
J. Leukocyte Biol.
72
:
621
-627.
2
Yeung, Y. G., E. R. Stanley.
2003
. Proteomic approaches to the analysis of early events in colony-stimulating factor-1 signal transduction.
Mol. Cell. Proteomics
2
:
1143
-1155.
3
Pixley, F. J., E. R. Stanley.
2004
. CSF-1 regulation of the wandering macrophage: complexity in action.
Trends Cell Biol.
14
:
628
-638.
4
Chitu, V., E. R. Stanley.
2006
. Colony-stimulating factor-1 in immunity and inflammation.
Curr. Opin. Immunol.
18
:
39
-48.
5
Cohen, P. E., L. Zhu, J. W. Pollard.
1997
. Absence of colony stimulating factor-1 in osteopetrotic (csfmop/csfmop) mice disrupts estrous cycles and ovulation.
Biol. Reprod.
56
:
110
-118.
6
Rohrschneider, L. R., R. P. Bourette, M. N. Lioubin, P. A. Algate, G. M. Myles, K. Carlberg.
1997
. Growth and differentiation signals regulated by the M-CSF receptor.
Mol. Reprod. Dev.
46
:
96
-103.
7
Dai, X. M., G. R. Ryan, A. J. Hapel, M. G. Dominguez, R. G. Russell, S. Kapp, V. Sylvestre, E. R. Stanley.
2002
. Targeted disruption of the mouse colony-stimulating factor 1 receptor gene results in osteopetrosis, mononuclear phagocyte deficiency, increased primitive progenitor cell frequencies, and reproductive defects.
Blood
99
:
111
-120.
8
Fixe, P., V. Praloran.
1998
. M-CSF: haematopoietic growth factor or inflammatory cytokine?.
Cytokine
10
:
32
-37.
9
Pandit, J., A. Bohm, J. Jancarik, R. Halenbeck, K. Koths, S. H. Kim.
1992
. Three-dimensional structure of dimeric human recombinant macrophage colony-stimulating factor.
Science
258
:
1358
-1362.
10
Taylor, E. W., A. L. Fear, A. Bohm, S. H. Kim, K. Koths.
1994
. Structure-function studies on recombinant human macrophage colony-stimulating factor (M-CSF).
J. Biol. Chem.
269
:
31171
-31177.
11
Glocker, M. O., B. Arbogast, J. Schreurs, M. L. Deinzer.
1993
. Assignment of the inter- and intramolecular disulfide linkages in recombinant human macrophage colony stimulating factor using fast atom bombardment mass spectrometry.
Biochemistry
32
:
482
-488.
12
Zhang, Y. H., X. Yan, C. S. Maier, M. I. Schimerlik, M. L. Deinzer.
2002
. Conformational analysis of intermediates involved in the in vitro folding pathways of recombinant human macrophage colony stimulating factor β by sulfhydryl group trapping and hydrogen/deuterium pulsed labeling.
Biochemistry
41
:
15495
-15504.
13
Falkenburg, J. H., M. A. Harrington, R. A. de Paus, W. K. Walsh, R. Daub, J. E. Landegent, H. E. Broxmeyer.
1991
. Differential transcriptional and posttranscriptional regulation of gene expression of the colony-stimulating factors by interleukin-1 and fetal bovine serum in murine fibroblasts.
Blood
78
:
658
-665.
14
Furman, I., C. Baudet, P. Brachet.
1996
. Differential expression of M-CSF, LIF, and TNF-α genes in normal and malignant rat glial cells: regulation by lipopolysaccharide and vitamin D.
J. Neurosci. Res.
46
:
360
-366.
15
Harrington, M., B. W. Konicek, X. L. Xia, A. Song.
1997
. Transcriptional regulation of the mouse CSF-1 gene.
Mol. Reprod. Dev.
46
:
39
-44.
16
Liu, M. Y., M. C. Wu.
1992
. Induction of human monocyte cell line U937 differentiation and CSF-1 production by phorbol ester.
Exp. Hematol.
20
:
974
-979.
17
Reisbach, G., J. Sindermann, J. P. Kremer, L. Hultner, H. Wolf, P. Dormer.
1989
. Macrophage colony-stimulating factor (CSF-1) is expressed by spontaneously outgrown EBV-B cell lines and activated normal B lymphocytes.
Blood
74
:
959
-964.
18
Hallet, M. M., V. Praloran, H. Vie, M. A. Peyrat, G. Wong, J. Witek-Giannotti, J. P. Soulillou, J. F. Moreau.
1991
. Macrophage colony-stimulating factor (CSF-1) gene expression in human T-lymphocyte clones.
Blood
77
:
780
-786.
19
Praloran, V., H. Gascan, S. Papin, S. Chevalier, M. Trossaert, M. C. Boursier.
1990
. Inducible production of macrophage colony-stimulating factor (CSF-1) by malignant and normal human T cells.
Leukemia
4
:
411
-414.
20
Irvine, K. M., C. J. Burns, A. F. Wilks, S. Su, D. A. Hume, M. J. Sweet.
2006
. A CSF-1 receptor kinase inhibitor targets effector functions and inhibits pro-inflammatory cytokine production from murine macrophage populations.
FASEB J.
20
:
1921
-1923.
21
Crowhurst, M. O., J. E. Layton, G. J. Lieschke.
2002
. Developmental biology of zebrafish myeloid cells.
Int. J. Dev. Biol.
46
:
483
-492.
22
Belosevic, M., P. C. Hanington, D. R. Barreda.
2006
. Development of goldfish macrophages in vitro.
Fish Shellfish Immunol.
20
:
152
-171.
23
Grabher, C., A. Cliffe, K. Miura, J. Hayflick, R. Pepperkok, P. Rorth, J. Wittbrodt.
2007
. Birth and life of tissue macrophages and their migration in embryogenesis and inflammation in medaka.
J. Leukocyte Biol.
81
:
263
-271.
24
Barreda, D. R., P. C. Hanington, J. L. Stafford, M. Belosevic.
2005
. A novel soluble form of the CSF-1 receptor inhibits proliferation of self-renewing macrophages of goldfish (Carassius auratus L.).
Dev. Comp. Immunol.
29
:
879
-894.
25
Honda, T., T. Nishizawa, M. Uenobe, C. Kohchi, A. Kuroda, M. Ototake, T. Nakanishi, Y. Yokomizo, Y. Takahashi, H. Inagawa, G. Soma.
2005
. Molecular cloning and expression analysis of a macrophage-colony stimulating factor receptor-like gene from rainbow trout, Oncorhynchus mykiss.
Mol. Immunol.
42
:
1
-8.
26
Roca, F. J., M. A. Sepulcre, G. Lopez-Castejon, J. Meseguer, V. Mulero.
2006
. The colony-stimulating factor-1 receptor is a specific marker of macrophages from the bony fish gilthead seabream.
Mol. Immunol.
43
:
1418
-1423.
27
Williams, H., S. Brenner, B. Venkatesh.
2002
. Identification and analysis of additional copies of the platelet-derived growth factor receptor and colony stimulating factor 1 receptor genes in fugu.
Gene
295
:
255
-264.
28
Hanington, P. C., T. Wang, C. J. Secombes, M. Belosevic.
2007
. Growth factors of lower vertebrates: characterization of goldfish (Carassius auratus L.) macrophage colony stimulating factor-1.
J. Biol. Chem.
282
:
31865
-31872.
29
Ganassin, R. C., N. C. Bols.
1998
. Development of a monocyte/macrophage-like cell line, RTS11, from rainbow trout spleen.
Fish Shellfish Immunol.
8
:
457
-476.
30
Wolf, K., M. C. Quimby.
1962
. Established eurythermic lines of fish cells in vitro.
Science
135
:
1065
-1066.
31
Lee, L. E., J. H. Clemons, D. G. Bechtel, S. J. Caldwell, K. B. Han, M. Pasitschniak-Arts, D. D. Mosser, N. C. Bols.
1993
. Development and characterization of a rainbow trout liver cell line expressing cytochrome P450-dependent monooxygenase activity.
Cell Biol. Toxicol.
9
:
279
-294.
32
Wang, T., C. J. Secombes.
2003
. Complete sequencing and expression of three complement components, C1r, C4, and C1 inhibitor, of the classical activation pathway of the complement system in rainbow trout Oncorhynchus mykiss.
Immunogenetics
55
:
615
-628.
33
Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, D. J. Lipman.
1997
. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.
Nucleic Acids Res.
25
:
3389
-3402.
34
Gasteiger, E., C. Hoogland, A. Gattiker, S. Duvaud, M. R. Wilkins, R. D. Appel, A. Bairoch.
2005
. Protein identification and analysis tools on the ExPASy server. J. M. Walker, ed.
The Proteomics Protocols Handbook
571
-607. Humana Press, Totowa, NJ.
35
Bendtsen, J. D., H. Nielsen, G. von Heijne, S. Brunak.
2004
. Improved prediction of signal peptides: SignalP 3.0.
J. Mol. Biol.
340
:
783
-795.
36
Hofmann, K., W. Stoffel.
1993
. TMbase: a database of membrane spanning proteins segments.
Biol. Chem. Hoppe-Seyler.
374
:
166
-166.
37
Campanella, J. J., L. Bitincka, J. Smalley.
2003
. MatGAT: an application that generates similarity/identity matrices using protein or DNA sequences.
BMC Bioinformatics
4
:
29
-32.
38
Letunic, I., R. R. Copley, B. Pils, S. Pinkert, J. Schultz, P. Bork.
2006
. SMART 5: domains in the context of genomes and networks.
Nucleic Acids Res.
34
:
D257
-D260.
39
Thompson, J. D., D. G. Higgins, T. J. Gibson.
1994
. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
Nucleic Acids Res.
22
:
4673
-4680.
40
Sinha, A. U., J. Meller.
2007
. Cinteny: flexible analysis and visualization of synteny and genome rearrangements in multiple organisms.
BMC Bioinformatics
8
:
82
-90.
41
Tamura, K., J. Dudley, M. Nei, S. Kumar.
2007
. MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0.
Mol. Biol. Evol.
24
:
1596
-1599.
42
Wang, T., N. Johnson, J. Zou, N. Bols, C. J. Secombes.
2004
. Sequencing and expression of the second allele of the interleukin-1β1 gene in rainbow trout (Oncorhynchus mykiss): identification of a novel SINE in the third intron.
Fish Shellfish Immunol.
16
:
335
-358.
43
Florea, L., G. Hartzell, Z. Zhang, G. M. Rubin, W. Miller.
1998
. A computer program for aligning a cDNA sequence with a genomic DNA sequence.
Genome Res.
8
:
967
-974.
44
Wang, T., J. W. Holland, A. Carrington, J. Zou, C. J. Secombes.
2007
. Molecular and functional characterization of IL-15 in rainbow trout Oncorhynchus mykiss: a potent inducer of IFN-γ expression in spleen leukocytes.
J. Immunol.
179
:
1475
-1488.
45
Wang, T., C. J. Secombes.
2008
. Rainbow trout suppressor of cytokine signalling (SOCS)-1, 2 and 3: molecular identification, expression, and modulation.
Mol. Immunol.
45
:
1449
-1457.
46
Wang, T., J. W. Holland, N. Bols, C. J. Secombes.
2005
. Cloning and expression of the first nonmammalian interleukin-11 gene in rainbow trout Oncorhynchus mykiss.
FEBS J.
272
:
1136
-1147.
47
Stafford, J. L., P. E. McLauchlan, C. J. Secombes, A. E. Ellis, M. Belosevic.
2001
. Generation of primary monocyte-like cultures from rainbow trout head kidney leukocytes.
Dev. Comp. Immunol.
25
:
447
-459.
48
Neumann, N. F., D. Barreda, M. Belosevic.
1998
. Production of a macrophage growth factor(s) by a goldfish macrophage cell line and macrophages derived from goldfish kidney leukocytes.
Dev. Comp. Immunol.
22
:
417
-432.
49
Jaillon, O., J. M. Aury, F. Brunet, J. L. Petit, N. Stange-Thomann, E. Mauceli, L. Bouneau, C. Fischer, C. Ozouf-Costaz, A. Bernot, et al
2004
. Genome duplication in the teleost fish Tetraodon nigroviridis reveals the early vertebrate proto-karyotype.
Nature
431
:
946
-957.
50
Zou, J., P. S. Grabowski, C. Cunningham, C. J. Secombes.
1999
. Molecular cloning of interleukin 1β from rainbow trout Oncorhynchus mykiss reveals no evidence of an ice cut site.
Cytokine
11
:
552
-560.
51
Bird, S., J. Zou, T. Wang, B. Munday, C. Cunningham, C. J. Secombes.
2002
. Evolution of interleukin-1β.
Cytokine Growth Factor Rev.
13
:
483
-502.
52
Liu, L., M. K. Callahan, D. Huang, R. M. Ransohoff.
2005
. Chemokine receptor CXCR3: an unexpected enigma.
Curr. Top. Dev. Biol.
68
:
149
-181.
53
Kwak, H. B., S. W. Lee, H. M. Jin, H. Ha, S. H. Lee, S. Takeshita, S. Tanaka, H. M. Kim, H. H. Kim, Z. H. Lee.
2005
. Monokine induced by interferon-γ is induced by receptor activator of nuclear factor κB ligand and is involved in osteoclast adhesion and migration.
Blood
105
:
2963
-2969.
54
Chang, M. X., B. J. Sun, P. Nie.
2007
. The first nonmammalian CXCR3 in a teleost fish: gene and expression in blood cells and central nervous system in the grass carp (Ctenopharyngodon idella).
Mol. Immunol.
44
:
1123
-1134.
55
Laing, K. J., N. Bols, C. J. Secombes.
2002
. A CXC chemokine sequence isolated from the rainbow trout Oncorhynchus mykiss resembles the closely related interferon-γ-inducible chemokines CXCL9, CXCL10 and CXCL11.
Eur. Cytokine Network
13
:
462
-473.
56
Zou, J., A. Carrington, B. Collet, J. M. Dijkstra, Y. Yoshiura, N. Bols, C. Secombes.
2005
. Identification and bioactivities of IFN-γ in rainbow trout Oncorhynchus mykiss: the first Th1-type cytokine characterized functionally in fish.
J. Immunol.
175
:
2484
-2494.